ai - CybAI news

20 Jul

Fine-tuning LLAMA 3 for Text Classification with Limited Resources

stp2y0 CommentsNewsai, coding, community, development, engineering, inclusive, machinelearning, python, software, tutorial

I recently needed to classify sentences for a particular use case at work. Remembering Jeremy Howard's Lesson 4: Getting started with NLP for absolute beginners, I first adapted his notebook to fine-tune DEBERTA. It worked, but not to my satisfaction, so I was curious what would happen if I used a LLM like LLAMA 3. The problem? Limited GPU resources. I only had access to a Tesla/Nvidia T4 instance. Research led me to QLORA. This tutorial on Fine tuning LLama 3 LLM for Text Classification of Stock Sentiment using QLoRA was particularly useful. To better understand the tutorial, I adapted…

19 Jul

Supercharging LLMs: RoT Fuses Language Models with Decision Tree Search to Boost Reasoning Power

stp2y0 CommentsNewsai, beginners, coding, community, datascience, development, engineering, inclusive, machinelearning, software

This is a Plain English Papers summary of a research paper called Supercharging LLMs: RoT Fuses Language Models with Decision Tree Search to Boost Reasoning Power. If you like these kinds of analysis, you should join AImodels.fyi or follow me on Twitter. Overview This paper explores a novel approach called "Reflection on Search Trees" (RoT) to enhance the capabilities of large language models (LLMs). RoT involves integrating tree search methods with LLMs to improve their reasoning and decision-making abilities. The paper presents the design and evaluation of the RoT system, demonstrating its effectiveness in outperforming traditional LLMs on various tasks.…

18 Jul

Multimodal RAG locally with CLIP and Llama3

stp2y0 CommentsNewsai, coding, community, development, engineering, inclusive, openai, rag, software, tutorial

With the recent release of GPT-4o and Gemini, multimodal has been a hot topic lately. Another one that has been on top of the lighting spot is Retrieval Augmented Generation (RAG) for the past year, but it was mostly focused on text. This tutorial will show you how to build a Multimodal RAG System. By using Multimodal RAG, you don’t have to use text only; you can use different types of data such as images, audio, videos, and text, of course. It’s also possible to return different kinds of data; just because you use text as input for your RAG…

18 Jul

OpenAI Slashes the Cost of Using Its AI With a “Mini” Model

stp2y0 CommentsGenAIai, artificial intelligence, openai

OpenAI today announced a cut-price “mini” model that it says will allow more companies and programs to tap into its artificial intelligence. The new model, called GPT-4o mini and available starting today, is 60 percent cheaper than OpenAI’s most inexpensive existing model while offering higher performance, the company says.OpenAI characterizes the move as part of an effort to make AI “as broadly accessible as possible,” but it also reflects growing competition among AI cloud providers as well as rising interest in small and free open source AI models. Meta is expected to debut the largest version of its very capable…

18 Jul

Microsoft releases iOS and Android apps for Designer, its AI-powered Canva competitor

stp2y0 CommentsNewsai, DALL-E, Designer, gear, image generation, microsoft, Microsoft Designer, news, openai

Microsoft has officially its platform for AI image generation. After a long preview phase, Designer is now available to most people with a Microsoft account. Designer can be used on the web in more than 80 languages, as a mobile app for iOS and Android, and as a Windows app. You can create a brand new visual from the ground up with AI, or use Designer to edit and tweak a picture you've already made. There are plenty of templates available to guide the creation of common image types, like a greeting card, smartphone wallpaper or a profile avatar. More…

18 Jul

Meta will reportedly withhold multimodal AI models from the EU amid regulatory uncertainty

stp2y0 CommentsNewsai, competition, eu, gear, meta, multimodal, news

Meta has decided to not offer its upcoming multimodal AI model and future versions to customers in the European Union citing a lack of clarity from European regulators, according to a report by Axios. The models in question are designed to process not only text but also images and audio, and power AI capabilities in Meta platforms as well as the company’s Ray-Ban smart glasses."We will release a multimodal Llama model over the coming months, but not in the EU due to the unpredictable nature of the European regulatory environment," Meta said in a statement to Axios.Meta’s move follows a…

17 Jul

OpenAI Touts New AI Safety Research. Critics Say It’s a Good Step, but Not Enough

stp2y0 CommentsGenAIai, artificial intelligence, openai, safety

OpenAI has faced opprobrium in recent months from those who suggest it may be rushing too quickly and recklessly to develop more powerful artificial intelligence. The company appears intent on showing it takes AI safety seriously. Today it showcased research that it says could help researchers scrutinize AI models even as they become more capable and useful.The new technique is one of several ideas related to AI safety that the company has touted in recent weeks. It involves having two AI models engage in a conversation that forces the more powerful one to be more transparent, or “legible,” with its…

17 Jul

How to Integrate AI into Your Node.js Application: A Step-by-Step Guide

stp2y0 CommentsNewsai, api, coding, community, development, engineering, gemini, inclusive, node, software

Introduction From being a fantasy to becoming a reality, Generative AI has proven to be a beneficial tool for many of us. It has boosted our overall productivity, automated repetitive tasks, and, in my case, created informative and educational content. That said, GenAI (Generative AI) still has a long way to go and shouldn't be fully relied upon for any given task. As a developer, you don't need to be an expert in AI or ML to build cool stuff. There are plenty of tools you can use to leverage the power of AI and integrate it into your projects.…

17 Jul

Qwen2 Technical Report

stp2y0 CommentsNewsai, beginners, coding, community, datascience, development, engineering, inclusive, machinelearning, software

This is a Plain English Papers summary of a research paper called Qwen2 Technical Report. If you like these kinds of analysis, you should subscribe to the AImodels.fyi newsletter or follow me on Twitter. Overview The provided paper is a technical report on the Qwen2 audio model. It covers the model's tokenizer, architecture, and other key technical details. The report aims to provide a comprehensive overview of the Qwen2 system for researchers and developers. Plain English Explanation The Qwen2 Technical Report outlines the technical details of the Qwen2 audio model. Qwen2 is a powerful machine learning model designed for various…

17 Jul

Apple, NVIDIA and Anthropic reportedly used YouTube transcripts without permission to train AI models

stp2y0 CommentsNewsai, anthropic, apple, dataset, gear, news, nvidia, openai, training data, youtube

Some of the world’s largest tech companies trained their AI models on a dataset that included transcripts of more than 173,000 YouTube videos without permission, a new investigation from Proof News has found. The dataset, which was created by a nonprofit company called EleutherAI, contains transcripts of YouTube videos from more than 48,000 channels and was used by Apple, NVIDIA and Anthropic among other companies. The findings of the investigation spotlight AI’s uncomfortable truth: the technology is largely built on the backs of data siphoned from creators without their consent or compensation.The dataset doesn’t include any videos or images from…